Adversarial Machine Learning in Image Classification: A Survey Toward the Defender’s Perspective

Deep Learning algorithms have achieved state-of-the-art performance for Image Classification. For this reason, they have been used even in security-critical applications, such as biometric recognition systems and self-driving cars. However, recent works have shown those algorithms, which can even surpass human capabilities, are vulnerable to adversarial examples. In Computer Vision, adversarial examples are images containing subtle perturbations generated by malicious optimization algorithms to fool classifiers. As an attempt to mitigate these vulnerabilities, numerous countermeasures have been proposed recently in the literature. However, devising an efficient defense mechanism has proven to be a difficult task, since many approaches demonstrated to be ineffective against adaptive attackers. Thus, this article aims to provide all readerships with a review of the latest research progress on Adversarial Machine Learning in Image Classification, nevertheless, with a defender’s perspective. This article introduces novel taxonomies for categorizing adversarial attacks and defenses, as well as discuss possible reasons regarding the existence of adversarial examples. In addition, relevant guidance is also provided to assist researchers when devising and evaluating defenses. Finally, based on the reviewed literature, this article suggests some promising paths for future research.

Three-Dimensional LiDAR Decoder Design for Autonomous Vehicles in Smart Cities

With the advancement of artificial intelligence, deep learning technology is applied in many fields. The autonomous car system is one of the most important application areas of artificial intelligence. LiDAR (Light Detection and Ranging) is one of the most critical components of self-driving cars. LiDAR can quickly scan the environment to obtain a large amount of high-precision three-dimensional depth information. Self-driving cars use LiDAR to reconstruct the three-dimensional environment. The autonomous car system can identify various situations in the vicinity through the information provided by LiDAR and choose a safer route. This paper is based on Velodyne HDL-64 LiDAR to decode data packets of LiDAR. The decoder we designed converts the information of the original data packet into X, Y, and Z point cloud data so that the autonomous vehicle can use the decoded information to reconstruct the three-dimensional environment and perform object detection and object classification. In order to prove the performance of the proposed LiDAR decoder, we use the standard original packets used for the comparison of experimental data, which are all taken from the Map GMU (George Mason University). The average decoding time of a frame is 7.678 milliseconds. Compared to other methods, the proposed LiDAR decoder has higher decoding speed and efficiency.

Future Blockchain Technology for Autonomous Applications/Autonomous Vehicle

The autonomous industry has rapidly grown for self-driving cars. The main purpose of autonomous industry is trying to give all types of security, privacy, secured traffic information to the self-driving cars. Blockchain is another newly established secured technology. The main aim of this technology is to provide more secured, convenient online transactions. By using this new technology, the autonomous industry can easily provide more suitable, safe, efficient transportation to the passengers and secured traffic information to the vehicles. This information can easily gather by the roadside units or by the passing vehicles. Also, the economical transactions can be possible more efficiently since blockchain technology allows peer-to-peer communications between nodes, and it also eliminates the need of the third party. This chapter proposes a concept of how the autonomous industry can provide more adequate, proper, and safe transportation with the help of blockchain. It also examines for the possibility that autonomous vehicles can become the future of transportation.

Protall (An Intelligent, multi-sensor, comprehensive obstacle avoidance system for automobiles and UAVs)

In this era of Artificial Intelligence and Automation, manufacturing and testing of self-driving cars and autonomous delivery of parcels to the desired location with the help of UAVs have a considerable amount of growth in the industry. This advancement in technology also raises safety issues due to the failure of sensors to detect the object or sometimes because of the dynamic environment. Protall (Protect-all) is an integrated solution for UAV and automobile vehicles to provide ultimate safety to both itself and its surroundings. With the strategic sensor integration and its intelligent processing, it aims to produce controlled output and hence, ensures to prevent any possible failures from occurring. The system constantly reacts to the environment and maintains comprehensive interaction with the user thus, enabling it to handle any dynamic situation and hence, it emerges as an optimal solution for Vehicles and UAVs.

Self-Driving Networks

This chapter presents a new vision of network operations, the self-driving network, that takes automation to the next level. This is not a description of existing work; rather, it is a challenge to dramatically rethink how we manage networks (or rather, how we do not manage networks). It draws upon an analogy with the development of self-driving cars and presents motivations for this effort. It then describes the technologies needed to implement this and an overall architecture of the system. As this endeavor will cause a major shift in network management, the chapter offers an evolutionary path to the end goal. Some of the consequences and human impacts of such a system are touched upon. The chapter concludes with some research topics and a final message. Key takeaways are that machine learning and feedback loops are fundamental to the solution; a key outcome is to build systems that are adaptive and predictive, for the benefit of users.